Adult stem cells reside in many tissues, where they replenish dying cells and repair tissue damage. In C. elegans and D. melanogaster germline stem cell (GSC) model systems, starvation leads to a decrease in GSC number, but some GSCs can survive. How these remaining GSCs are preserved during the prolonged absence of nutrients is unclear. One prevailing hypothesis is that GSCs become more quiescent and thus more resilient to nutrient stress. To gain insight into the biological basis of this phenomenon, we explored the effects of starvation in the Drosophila testis, a model system in which GSCs can be clearly identified and distinguished from their spermatogonia (SG) daughters. We discovered that starvation does not affect GSC proliferation compared with the fed condition. Instead starvation has a much more striking effect in the SG compartment where we detected an increase in the rate of cell death. This observation accounts for the dramatic downsizing of the tissue as a whole despite only a minor decrease in the number of active GSCs. We hypothesize that starvation induces cell death in SGs, which decreases nutrient demand and allows GSCs to be maintained in active proliferation. To test our hypothesis, we will elucidate the mechanism of how SGs die under fed and starved conditions. We will then explore whether SG death is required for GSC maintenance during starvation. If our hypothesis is true, it would provide a novel paradigm by which stem cells are preserved under nutrient stress: at the expense of differentiating cells. While this reaffirms the importance of protecting stem cells in a tissue, it challenges the belief that the tissue adaptation to stress is accomplished primarily through changes at the stem cell level.